L.S. Ibels

Fasting serum triglyceride and cholesterol measurements, and lipoprotein characterization by ultracentrifugation, were performed in four groups of patients with chronic renal disease (uraemic, short- and long-term haemodialysis and renal transplant recipients) and the results compared with those obtained from age- and sex-matched control subjects. Basal insulin and growth hormone levels, and serum creatinine and albumin concentrations were measured in, and detailed dietary histories taken from patients in each group. The predominant lipid abnormalities were hypertriglyceridaemia and increased very low density lipoproteins (type IV hyperlipoproteinaemia) in both uraemic and haemodialysis patients. Following renal transplantation, a different pattern of hyperlipidaemia was found. Hypercholesterolaemia was more common and hypertriglyceridaemia less common than in the uraemic and haemodialysis group. The lipoprotein abnormalities were increased low density and/or very low density lipoproteins, with types IIa IIb and IV hyperlipoproteinaemia occurring equally frequently. In uraemic and haemodialysis patients, the proportion of carbohydrate in the diet was high, and may have played a role in the genesis of hypertriglyceridaemia. There was a positive correlation between relative body weight and serum triglyceride in the long-term dialysis group. In renal allograft recipients hypertriglyceridaemia could be attributed, at least in part, to obesity, prednisone dosage and the degree of impairment of graft function. The aetiology of hypercholesterolaemia in the transplant recipients was unclear. Neither basal insulin nor growth hormone levels were elevated in any patient group. Uraemic hypertriglyceridaemia is a clearly defined and well documented metabolic abnormality which is not corrected by dialysis. Post-transplantation hyperlipidaemia however, is a condition of variable presentation and multifactorial aetiology.

The prevalence of clinical and sub-clinical occlusive arterial disease and of risk factors implicated in the pathogenesis of arteriosclerosis was assessed in 21 patients with chronic renal failure, 27 on maintenance haemodialysis and 51 renal allograft recipients. Clinical occlusive arterial disease was present in 27 patients, and sub-clinical arterial disease in 34. Myocardial infarction, cerebral thrombosis and lower limh arterial thrombosis had occurred only in the transplant recipients; these patients had, however, been followed for a longer period of time than the other two groups. In the allograft recipients, the cumulative incidence of any occlusive arterial disease was 416 per 1000, and that of coronary heart disease was 267 per 1000 at six years. Hypertension was present in 76 per cent of patients prior to renal replacement therapy. Following institution of definitive therapy, hypertension was of shorter duration and less common in haemodialysis patients than in renal transplant recipients. Uraemic and haemodialysis patients with occlusive arterial disease had required antihypertensive medication for significantly longer than those free of arterial disease. Transplant recipients with hypertension had a greater mean serum creatinine, were receiving a larger maintenance dosage of corticosteroids and less frequently had undergone prior bilateral nephrectomy than those transplant patients without hypertension. Serum lipid levels were elevated in 62 per cent of patients. In the uraemic and haemodialysis patients hypertriglyceridaemia was the predominant abnormality while in the transplant recipients combined hypertriglyceridaemia/hyperchole-sterolaemia was more frequent. Despite regular aluminium hydroxide therapy 81 per cent of uraemic and haemodialysis patients had a calcium × phosphate product higher than normal. Arterial and/or soft tissue calcification as demonstrable in 20–38 per cent of patients within each group, but could not be related to the calcium × phosphate product or radiographic evidence of hyperparathyroidism. Glucose intolerance was present in 71 per cent of the uraemic and haemodialysis patients and 33 per cent of the transplant recipients. Hyperuricaemia, cigarette smoking, obesity and a sedentary existence were also prevalent. The majority of patients had several risk factors implicated in the pathogenesis of arteriosclerosis. Occluaive arterial disease is a major problem in patients with end stage renal disease, being no less common after transplantation than with long-term maintenance dialysis. The aetiology is multifactorial.

In a series of 325 recipients of cadaveric renal transplants sudden occlusive arterial disease was found to be responsible for 12% of deaths. Acute myocardial infarction (9%) occurred 25 times more than expected in the normal population and cerebral thrombosis (3%) 300 times more. The greatest loss was in the initial three-month period after transplantation. Patients with renal failure due to essential hypertension were especially at risk, accounting for six of the 12 deaths.

The dietary protein intake (DPI) of 766 patients (aged 7 to 88 yr) was determined from 24-h urinary urea and protein excretion by urea kinetic modelling. Five hundred sixty-five patients had a normal serum creatinine concentration, and of these 565, 385 patients had no dietary modification advised and 180 were advised to follow a low-protein diet. The remaining 201 patients had an increased serum creatinine concentration; 148 of these 201 patients had been advised to restrict their DPI. Patients with a normal serum creatinine concentration who had no dietary restriction had a significantly higher DPI than those advised to restrict their protein intake (1.08 +/- 0.01 versus 0.96 +/- 0.02 g/kg per day (mean +/- SEM), P < 0.01). Similarly, patients with abnormal renal function who were advised to follow a low-protein diet had a reduced DPI (0.93 +/- 0.01 versus 0.87 +/- 0.02 g/kg per day; P < 0.05). A lower DPI correlated with level of renal dysfunction, independent of dietary advice (P < 0.0001). In the overall population, DPI correlated with body mass index (BMI; P < 0.0001) and serum albumin (P < 0.0001), and inverse correlations were evident between age (P < 0.0001), blood glucose level (P < 0.01), serum cholesterol level (P < 0.0001), and triglyceride levels (P < 0.0001) independently of renal function. Fifty-two patients were assessed within the 3 months before the commencement of dialysis, and 47 were reassessed within 3 months after the commencement of dialysis. Despite advice regarding an increase in dietary protein after the commencement of dialysis, this increase failed to occur within the 3 months of commencement of dialytic therapy (0.79 +/- 0.04 versus 0.82 +/- 0.03 g/kg per day); P = 0.64). However, 6 to 9 months after the commencement of dialysis, a significant increase in protein intake was evident (1.04 +/- 0.04 g/kg per day; P < 0.005 versus both prior measurements). Hence a low DPI in renal impairment occurs independently of dietary advice, but compliance with such advice is evident because patients advised to consume a low-protein diet had significantly lower protein intake than did patients receiving no dietary advice. Adaptation to a high-protein diet after instigation of dialysis is unsuccessful in the short term, irrespective of whether or not advice is given regarding a low-protein diet before dialysis is initiated. However, 6 to 9 months after the commencement of dialysis, a significant increase in protein intake occurs, which in the hemodialysis population correlates with dialysis delivery.

In order to assess long-term nutritional adequacy, 154 patients on maintenance dialysis (78 on hemodialysis (HD), 76 on continuous ambulatory peritoneal dialysis (CAPD)) underwent measurement of total body nitrogen (TBN) with concurrent recording of dietary history, anthropometrics, and serum albumin. Seventy-one patients were reassessed 23.3 +/- 2.2 (5 to 76) months later. In cross-sectional analyses, anthropometric measurements and dietary intake remained stable over time in all patients. However, a significant fall in TBN occurred in the HD population with increasing time on dialysis (P < 0.05). In the prospective analyses, CAPD patients (N = 26) had a significant increase in TBN (P < 0.02). In contrast, longitudinal measurements of TBN in HD patients (N = 36) tended to fall but did not reach significance (P = 0.18). TBN correlated with total caloric intake estimated from the dietary history (P < 0.05), but not with estimated protein intake. During follow-up, 38 patients died. These patients were older (P < 0.05), and in the CAPD population, they had been on dialysis for a longer time (P < 0.05). Those who died had a lower TBN expressed both as grams per kilogram lean body mass (P < 0.005) and as the nitrogen index (P < 0.05). The probability of death within 12 months in the patients with a nitrogen index (ratio of the measured nitrogen to the predicted nitrogen for a sex-, age-, and height-matched control) less than 80% of the predicted normal value was 48%. The relative risk of death in this population was 4.1.(ABSTRACT TRUNCATED AT 250 WORDS)